Clinical resistance to chemotherapeutic drugs is a major impediment to the successful treatment of human neoplasia. Though a great deal of research on model cell lines and tumors has allowed an understanding of the determinants of response and the mechanisms of resistance to a variety of useful drugs, there is need to expand existing information and to assess its significance to the clinical setting. Thus, it is pertinent to examine the extent and impact of genetic variation on parameters of drug response in human tumor cells of relevant histological origin. Toward these goals, we intend to study molecular and biochemical variation in expression of thymidylate synthase (TS) in colon tumor cells. Since TS is a target for fluoropyrimidine anti-metabolites, which are widely used in treating carcinomas of the ovary, breast, and gastrointestinal tract, it is expected that variation in parameters related to the enzyme's role as a target will profoundly affect sensitivity to these drugs. We will focus on a panel of human colon tumor cell lines that represent a sampling of the inter- and intra-tumoral phenotypic heterogeneity characteristic of colon cancer. Parameters to be measured in these lines include: synthesis and turnover of the active anti-metabolite (5-fluoro-2'-deoxyuridylate); levels of folate cofactor; structure and concentration of TS protein and mRNA; and copy-number and structure of TS genes. Variations in these phenotypes will be correlated with growth sensitivities to the fluoropyrimidine analog 5-fluoro-2'-deoxyuridine. The study of variation in TS expression and its function as a drug target is important in elucidating factors governing drug response and may lead to the identification of markers that could be useful in the pre-clinical prediction of therapeutic response in many patients. In all, these studies should generate information promoting the development of a more rational basis for improving the chemotherapy of cancer.